There are nearly 200,000 in-hospital cardiac arrests (IHCA) in the U.S each year, but less than a quarter of patients survive to discharge. When patients do survive, significant neurological impairment is common. Despite the resources and opportunity to intervene early when cardiac arrests occur in hospitals, IHCA incidence and outcomes vary widely from one hospital to the next. Evidence suggests that differences in the recognition and response to early warning signs contribute to the variation in IHCA incidence. Likewise, important processes of care like time to chest compressions and defibrillation are key to survival and good neurological function when IHCA does occur. There is little evidence, however, to help explain why processes of care and patient outcomes vary so much from hospital to hospital, and importantly, what could be done about it. Nurses are the primary clinical surveillance system in the hospital: they are at the bedside 24 hours a day every day with direct knowledge of changes in patient condition, and they are often the first on the scene of an IHCA where they initiate and coordinate the activities of others, including physicians, to save a patient's life. Thus, nurses are ideally positioned to identify clinical deterioration and mobilize life-saving interventions. Prior research points to nursing factors, including nurse staffing, nurs education, nursing skill mix, and the nurse work environment, as vital to nurses' ability to identiy and respond to deterioration in patient condition in order to prevent poor outcomes. The relationships between nursing and the incidence, overall survival, and neurological outcomes after IHCA, however, are unknown. We address this gap by taking advantage of multiple large and unique databases: The American Heart Association's Get With The Guidelines- Resuscitation (GWTG-R) clinical registry (the gold standard for IHCA research) and the 2015 RN4CAST-US (R01-NR014855) study. The clinical registry incorporates uniformly reported and precisely defined clinical variables developed by international experts. Using unique hospital identifiers to link patient outcomes and care processes from GWTG-R with measures of nursing from RN4CAST-US, we will identify specific, actionable elements of nursing that influence IHCA patient outcomes. Because we have new 2015 data for cross-sectional analyses as well as repeated cross-sections (2006, 2015) of hospitals and their patients at two points in time, our study will allow us to learn about causal relationships and determine whether changes in nursing factors over time lead to changes in IHCA patient outcomes. We can also evaluate whether more effective clinical care (more monitored/witnessed events, shorter time to chest compressions and defibrillation) is an important mechanism through which better nurse work environments have their effect. This unique combination of data over time and detailed process of care and patient outcomes information will allow us to clarify causal linkages between nursing and IHCA outcomes specifically, while advancing our understanding of how changing nurse work environments can yield better patient outcomes through better clinical practices.